The overall objectives are to design a family-based adolescent alcohol misuse prevention intervention and conduct a reduced scale feasibility study. Methodological aims are to assess recruitment and retention rates for the experimental conditions. Substantive aims are to conduct pilot studies of conceptual models of adolescent alcohol use and misuse. Based on the results of those tests, a family-based intervention directed at altering parental norms and behavior regarding adolescent alcohol use and misuse will be designed and pilot tested. The specific aims of the pilot test are to: 1) test, in cross-sectional and longitudinal studies, the predictive validity of components of a conceptual model of parent and peer group influences on adolescent alcohol use and misuse; 2)pilot test the effectiveness of the intervention in changing parents' information, attitudes and behavior, and their children's alcohol use/misuse intentions and behavior, and 3)pilot test the effectiveness of the parent-directed intervention using procedures designed to maximize intervention effectiveness with procedures generalizable to use as a component of a school-based intervention. Subjects will be 600 students, in the fourth grade in year one and their parents. All target students will be tested in their classrooms, which will yield additional data on approximately 340 non-target peers. One third of the 600 families will be randomly assigned to each of three conditions: maximum treatment, generalizable treatment, and control. Half of each of those groups will be randomly assigned to pretest to no-pretest conditions, and half of each of these subgroups will be assigned to demand instructions before and after conditions. One third of each of these groups will be randomly assigned to each of three incentive conditions (payment, drawing, none). The maximum treatment condition parents will receive three visits and discussion concerning the consequences of adolescent alcohol use, the relationship of parental alcohol and other drug use to adolescent alcohol use/misuse, and parenting skills designed to prevent adolescent alcohol misuse. The visits will be followed by phone contacts mailings, or interviewer availability to answer questions. The generalizable treatment condition parents will receive the same information in mailed pamphlets, but no home visits, follow-up phone calls, or interviewer availability to answer questions. The control group will receive no information. All students will be pretested in their classrooms. Half of the parents in each condition will be pretested with home interviews prior to the intervention. All students will be posttested in their classroom, and all parents will be posttested via home interviews once each year following the intervention. The data analyses will consist of repeated measures analysis of variance and interactive testing of covariance structure models of parental practices and adolescent alcohol use/misuse. Child gender effects will be routinely assessed. Ethnicity effects will be assessed insofar as possible with the pilot study sample size.